Artificial-Intelligence-Based Prediction of Clinical Events among Hemodialysis Patients Using Non-Contact Sensor Data

Advancements in Breathomics: Special Focus on Electrochemical Sensing and AI for Chronic Disease Diagnosis and Monitoring

This Review examines the potential of breathomics in enhancing disease monitoring and diagnostic precision when integrated with artificial intelligence (AI) and electrochemical sensing techniques. It discusses breathomics' potential for early and noninvasive disease diagnosis with a focus on chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), and lung cancer, which have been well studied in the context of VOC association with diseases. The noninvasive nature of exhaled breath analysis can be advantageous compared to traditional diagnostic methods for CKD, which often rely on blood and urine testing. VOC analysis can enhance spirometry and imaging methods used in COPD diagnosis, providing a more comprehensive picture of the disease's progression. Breathomics could also provide a less intrusive and potentially earlier diagnostic approach for lung cancer, which is now dependent on imaging and biopsy. The combination of breathomics, electrochemical sensing, and AI could lead to more personalized and successful treatment plans for chronic illnesses using AI algorithms to decipher complicated VOC patterns. This Review assesses the viability and effectiveness of combining breathomics with electrochemical sensors and artificial intelligence by synthesizing recent research findings and technological developments.

Feedback control in hemodialysis

A number of systems of feedback control during dialysis have been developed, which have the shared characteristic of prospectively measuring physiological parameters and then automatically altering dialysis parameters in real time according to a pre-specified dialysis prescription. These include feedback systems aimed at reducing intradialytic hypotension based on relative blood volume monitoring linked to adjustments in ultrafiltration and dialysate conductivity, and blood temperature monitoring linked to alterations in dialysate temperature. Feedback systems also exist that manipulate sodium balance during dialysis by assessing and adjusting dialysate conductivity. In this review article, we discuss the rationale for automated feedback systems during dialysis, describe how the different feedback systems work, and provide a review of the current evidence on their clinical effectiveness.

Deep Learning on Electrocardiograms for Prediction of In-hospital Intradialytic Hypotension in Patients with ESKD

Intradialytic hypotension is common in patients who are on hemodialysis. We applied deep learning techniques to ECGs to predict patients at risk of IDH. The performance of the model was good with an AUC of 0.763 and AUPRC of 0.35.

Early prediction of hemodialysis complications employing ensemble techniques

Background and objectives

Hemodialysis complications remain a critical threat among dialysis patients. They result in sudden termination of the session which impacts the efficiency of dialysis. As intra-dialytic complications are the result of the interplay of multiple factors, artificial intelligence can aid in their early prediction. This research aims to compare different machine learning tools for the early prediction of the most frequent hemodialysis complications with high performance, using the fewest predictors for easier practical implementation.

Methods

Fifty different variables were recorded during 6000 hemodialysis sessions performed in a regional dialysis unit in Egypt. The filter technique was used to extract the most relevant features. Then, five individual classifiers and three ensemble approaches were implemented to predict the occurrence of intra-dialytic complications. Different subsets of 25, 12 and 6 from the 50 collected features were tested.

Results

Random forest yielded the highest accuracy of 98% with the least training time using 12 features in a balanced dataset, while the gradient boosting allowed obtaining the highest F1-score of 94%, 92%, and 78% in the prediction of hypotension, hypertension, and dyspnea, respectively, in imbalanced datasets.

Conclusion

Applying different machine learning algorithms to big datasets can improve accuracy, reduce training time and model complexity allowing simple implementation in clinical practice. Our models can help nephrologists predict and possibly prevent dialysis complications.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12938-022-01044-0.

Deep Learning Model for Predicting Intradialytic Hypotension Without Privacy Infringement: A Retrospective Two-Center Study

Objective

Previously developed Intradialytic hypotension (IDH) prediction models utilize clinical variables with potential privacy protection issues. We developed an IDH prediction model using minimal variables, without the risk of privacy infringement.

Methods

Unidentifiable data from 63,640 hemodialysis sessions (26,746 of 79 patients for internal validation, 36,894 of 255 patients for external validation) from two Korean hospital hemodialysis databases were finally analyzed, using three IDH definitions: (1) systolic blood pressure (SBP) nadir <90 mmHg (Nadir90); (2) SBP decrease ≥20 mmHg from baseline (Fall20); and (3) SBP decrease ≥20 mmHg and/or mean arterial pressure decrease ≥10 mmHg (Fall20/MAP10). The developed models use 30 min information to predict an IDH event in the following 10 min window. Area under the receiver operating characteristic curves (AUROCs) and precision-recall curves were used to compare machine learning and deep learning models by logistic regression, XGBoost, and convolutional neural networks.

Results

Among 344,714 segments, 9,154 (2.7%), 134,988 (39.2%), and 149,674 (43.4%) IDH events occurred according to three different IDH definitions (Nadir90, Fall20, and Fall20/MAP10, respectively). Compared with models including logistic regression, random forest, and XGBoost, the deep learning model achieved the best performance in predicting IDH (AUROCs: Nadir90, 0.905; Fall20, 0.864; Fall20/MAP10, 0.863) only using measurements from hemodialysis machine during dialysis session.

Conclusions

The deep learning model performed well only using monitoring measurement of hemodialysis machine in predicting IDH without any personal information that could risk privacy infringement.

Recent Trends in AI-Based Intelligent Sensing

In recent years, intelligent sensing has gained significant attention because of its autonomous decision-making ability to solve complex problems. Today, smart sensors complement and enhance the capabilities of human beings and have been widely embraced in numerous application areas. Artificial intelligence (AI) has made astounding growth in domains of natural language processing, machine learning (ML), and computer vision. The methods based on AI enable a computer to learn and monitor activities by sensing the source of information in a real-time environment. The combination of these two technologies provides a promising solution in intelligent sensing. This survey provides a comprehensive summary of recent research on AI-based algorithms for intelligent sensing. This work also presents a comparative analysis of algorithms, models, influential parameters, available datasets, applications and projects in the area of intelligent sensing. Furthermore, we present a taxonomy of AI models along with the cutting edge approaches. Finally, we highlight challenges and open issues, followed by the future research directions pertaining to this exciting and fast-moving field.

From reverse innovation to global innovation in animal health: A review

Reverse innovation refers to learning from or diffusion of innovations developed in low income settings and further translated to industrialized countries. There is lack of consensus regarding terminology, but the idea that innovations in low-income countries are promising for adoption in high-income contexts is not new. However, in healthcare literature globally, the vast majority of publications referring to 'disruptive innovation' were published in the last ten years. To assess the potential of innovative developments and technologies for improving animal health, we initiated a literature review in 2020. We used a combined approach, incorporating targeted searching in PubMed using a key word algorithm with a snowball technique, to identify 120 relevant publications and extract data for qualitative coding. Heterogeneity of articles precluded meta-analysis, quality scoring and risk of bias analysis. We can distinguish technical innovations like new digital devices, diagnostic tests and procedures, and social innovations of intersectoral cooperation. We profile two case studies to describe potential global innovations: an integrated surveillance and response system in Somali Regional State, Ethiopia and a blockchain secured One Health intervention to optimally provide post-exposure prophylaxis for rabies exposed people in West Africa. Innovation follows no borders and can also occur in low-income settings, under constraints of cost, lack of services and infrastructure. Lower administrative and legal barriers may contribute to produce innovations that would not be possible under conditions of high density of regulation. We recommend using the term global innovation, which highlights those emanating from international partnership to solve problems of global implications.

Volume Balance in Chronic Kidney Disease: Evaluation Methodologies and Innovation Opportunities

BACKGROUND

Patients affected by chronic kidney disease are at a risk of cardiovascular morbidity and mortality. Body fluids unbalance is one of the main characteristics of this condition, as fluid overload is highly prevalent in patients affected by the cardiorenal syndrome.

SUMMARY

We describe the state of the art and new insights into body volume evaluation. The mechanisms behind fluid balance are often complex, mainly because of the interplay of multiple regulatory systems. Consequently, its management may be challenging in clinical practice and even more so out-of-hospital. Availability of novel technologies offer new opportunities to improve the quality of care and patients' outcome. Development and validation of new technologies could provide new tools to reduce costs for the healthcare system, promote personalized medicine, and boost home care. Due to the current COVID-19 pandemic, a proper monitoring of chronic patients suffering from fluid unbalances is extremely relevant. Key Message: We discuss the main mechanisms responsible for fluid overload in different clinical contexts, including hemodialysis, peritoneal dialysis, and heart failure, emphasizing the potential impact provided by the implementation of the new technologies.

Deep Learning Model for Real-Time Prediction of Intradialytic Hypotension

BACKGROUND AND OBJECTIVES

Intradialytic hypotension has high clinical significance. However, predicting it using conventional statistical models may be difficult because several factors have interactive and complex effects on the risk. Herein, we applied a deep learning model (recurrent neural network) to predict the risk of intradialytic hypotension using a timestamp-bearing dataset.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

We obtained 261,647 hemodialysis sessions with 1,600,531 independent timestamps (i.e., time-varying vital signs) and randomly divided them into training (70%), validation (5%), calibration (5%), and testing (20%) sets. Intradialytic hypotension was defined when nadir systolic BP was <90 mm Hg (termed intradialytic hypotension 1) or when a decrease in systolic BP ≥20 mm Hg and/or a decrease in mean arterial pressure ≥10 mm Hg on the basis of the initial BPs (termed intradialytic hypotension 2) or prediction time BPs (termed intradialytic hypotension 3) occurred within 1 hour. The area under the receiver operating characteristic curves, the area under the precision-recall curves, and F1 scores obtained using the recurrent neural network model were compared with those obtained using multilayer perceptron, Light Gradient Boosting Machine, and logistic regression models.

RESULTS

The recurrent neural network model for predicting intradialytic hypotension 1 achieved an area under the receiver operating characteristic curve of 0.94 (95% confidence intervals, 0.94 to 0.94), which was higher than those obtained using the other models (P<0.001). The recurrent neural network model for predicting intradialytic hypotension 2 and intradialytic hypotension 3 achieved area under the receiver operating characteristic curves of 0.87 (interquartile range, 0.87-0.87) and 0.79 (interquartile range, 0.79-0.79), respectively, which were also higher than those obtained using the other models (P≤0.001). The area under the precision-recall curve and F1 score were higher using the recurrent neural network model than they were using the other models. The recurrent neural network models for intradialytic hypotension were highly calibrated.

CONCLUSIONS

Our deep learning model can be used to predict the real-time risk of intradialytic hypotension.

Edge-Event-Triggered Synchronization for Multi-Agent Systems with Nonlinear Controller Outputs

This paper addresses the synchronization problem of multi-agent systems with nonlinear controller outputs via event-triggered control, in which the combined edge state information is utilized, and all controller outputs are nonlinear to describe their inherent nonlinear characteristics and the effects of data transmission in digital communication networks. First, an edge-event-triggered policy is proposed to implement intermittent controller updates without Zeno behavior. Then, an edge-self-triggered solution is further investigated to achieve discontinuous monitoring of sensors. Compared with the previous event-triggered mechanisms, our policy design considers the controller output nonlinearities. Furthermore, the system’s inherent nonlinear characteristics and networked data transmission effects are combined in a unified framework. Numerical simulations demonstrate the effectiveness of our theoretical results.

Unobtrusive Health Monitoring in Private Spaces: The Smart Vehicle

Unobtrusive in-vehicle health monitoring has the potential to use the driving time to perform regular medical check-ups. This work intends to provide a guide to currently proposed sensor systems for in-vehicle monitoring and to answer, in particular, the questions: (1) Which sensors are suitable for in-vehicle data collection? (2) Where should the sensors be placed? (3) Which biosignals or vital signs can be monitored in the vehicle? (4) Which purposes can be supported with the health data? We reviewed retrospective literature systematically and summarized the up-to-date research on leveraging sensor technology for unobtrusive in-vehicle health monitoring. PubMed, IEEE Xplore, and Scopus delivered 959 articles. We firstly screened titles and abstracts for relevance. Thereafter, we assessed the entire articles. Finally, 46 papers were included and analyzed. A guide is provided to the currently proposed sensor systems. Through this guide, potential sensor information can be derived from the biomedical data needed for respective purposes. The suggested locations for the corresponding sensors are also linked. Fifteen types of sensors were found. Driver-centered locations, such as steering wheel, car seat, and windscreen, are frequently used for mounting unobtrusive sensors, through which some typical biosignals like heart rate and respiration rate are measured. To date, most research focuses on sensor technology development, and most application-driven research aims at driving safety. Health-oriented research on the medical use of sensor-derived physiological parameters is still of interest.

Use of Internet of Things With Data Prediction on Healthcare Environments

Internet of Things (IoT) is a constantly growing paradigm that promises to revolutionize healthcare applications and could be associated with several other techniques. Data prediction is another widely used paradigm, where data captured over time is analyzed in order to identify and predict problematic situations that may happen in the future. After research, no surveys that address IoT combined with data prediction in healthcare area exist in the literature. In this context, this work presents a systematic literature review on Internet of Things applied to healthcare area with a focus on data prediction, presenting twenty-three papers about this theme as results, as well as a comparative analysis between them. The main contribution for literature is a taxonomy for IoT systems with data prediction applied to healthcare. Finally, this article presents the possibilities and challenges of exploration in the study area, showing the existing gaps for future approaches.

A Joint Evaluation of Impaired Cardiac Sympathetic Responses and Malnutrition-Inflammation Cachexia for Mortality Risks in Hemodialysis Patients

Background: Cardiac sympathetic response (CSR) and malnutrition-inflammation syndrome (MIS) score are validated assessment tools for patients' health condition. We aim to evaluate the joint effect of CSR and MIS on all-cause and cardiovascular (CV) mortality in patients with hemodialysis (HD).

Methods: Changes in normalized low frequency (ΔnLF) during HD were utilized for quantification of CSR. Unadjusted and adjusted hazard ratios (aHRs) of mortality risks were analyzed in different groups of ΔnLF and MIS score.

Results: In multivariate analysis, higher ΔnLF was related to all-cause, CV and sudden cardiac deaths [aHR: 0.78 (95% confidence interval (CI): 0.72–0.85), 0.78 (95% CI: 0.70–0.87), and 0.74 (95% CI: 0.63–0.87), respectively]. Higher MIS score was associated with incremental risks of all-cause, CV and sudden cardiac deaths [aHR: 1.36 (95% CI: 1.13–1.63), 1.33 (95% CI: 1.06 – 1.38), and 1.50 (95% CI: 1.07–2.11), respectively]. Patients with combined lower ΔnLF (≤6.8 nu) and higher MIS score were at the greatest risk of all-cause and CV mortality [aHR: 5.64 (95% CI: 1.14–18.09) and 5.86 (95% CI: 1.64–13.65), respectively].

Conclusion: Our data indicate a joint evaluation of CSR and MIS score to identify patients at high risk of death is more comprehensive and convincing. Considering the extremely high prevalence of cardiac autonomic neuropathy and malnutrition-inflammation cachexia in HD population, a non-invasive monitoring system composed of CSR analyzer and MIS score calculator should be developed in the artificial intelligence-based prediction of clinical events.

Review-Microwave Radar Sensing Systems for Search and Rescue Purposes

This paper presents a survey of recent developments using Doppler radar sensor in searching and locating an alive person under debris or behind a wall. Locating a human and detecting the vital signs such as breathing rate and heartbeat using a microwave sensor is a non-invasive technique. Recently, many hardware structures, signal processing approaches, and integrated systems have been introduced by researchers in this field. The purpose is to enhance the accuracy of vital signs' detection and location detection and reduce energy consumption. This work concentrates on the representative research on sensing systems that can find alive people under rubble when an earthquake or other disasters occur. In this paper, various operating principles and system architectures for finding survivors using the microwave radar sensors are reviewed. A comparison between these systems is also discussed.

Flocking of Multi-Agent System with Nonlinear Dynamics via Distributed Event-Triggered Control

In this paper, a distributed event-triggered control strategy is proposed to investigate aflocking problem in a multi-agent system with Lipschitz nonlinear dynamics, where triggeringconditions are proposed to determine the instants to update the controller. A distributedevent-triggered control law with bounded action function is proposed for free flocking. It is provedthat the designed event-triggered controller ensures a group of agents reach stable flocking motionwhile preserving connectivity of the communication network. Lastly, simulations are provided toverify the effectiveness of the theoretical results.